I’m trying to use the reana-client restart
tool to re-run the final step of my workflow after making some updates, and having some trouble. The final step (as encoded in the yadage workflow.yml) is called interpretation_stage
:
- name: interpretation_stage
dependencies: [make_dict_stage, skimming_stage_mc16a, init]
scheduler:
scheduler_type: singlestep-stage
parameters:
input_dict: {step: make_dict_stage, output: python_dict}
monojet_conf: {step: skimming_stage_mc16a, output: monojet_conf}
makeLimitScript: {step: init, output: makeLimitScript}
outputdir: '{workdir}/Interpretation_output'
step: {$ref: steps.yml#/interpretation_stage}
Here is my reana.yaml
:
version: 0.6.0
inputs:
parameters:
run_num: '311454'
DAOD_mc16a: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/DAODs/mc16a/mc16_13TeV.311454.PowhegPy8EG_A14N23LO_DMA_1_10_gq0p25.deriv.DAOD_EXOT5.e7368_a875_r9364_p3712/DAOD_EXOT5.17576479._000002.pool.root.1'
DAOD_mc16d: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/DAODs/mc16d/mc16_13TeV.311454.PowhegPy8EG_A14N23LO_DMA_1_10_gq0p25.deriv.DAOD_EXOT5.e7368_a875_r10201_p3712/DAOD_EXOT5.17576865._000001.pool.root.1'
DAOD_mc16e: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/DAODs/mc16e/mc16_13TeV.311454.PowhegPy8EG_A14N23LO_DMA_1_10_gq0p25.deriv.DAOD_EXOT5.e7368_a875_r10724_p3712/DAOD_EXOT5.17576468._000001.pool.root.1'
#xSec_dir: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/xSection'
#antiSF_dir: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/antiSF'
makeLimitScript: 'root://eosuser.cern.ch//eos/project/r/recast/atlas/ANA-EXOT-2018-06/scripts/makeLimitFile.py'
workflow:
type: yadage
file: workflow.yml
outputs:
files:
- interpretation_stage/Interpretation_output/LimitDir/asymptotics/limit_CL95.txt
I tried restarting the step from the workflow named monojet.26
(see attached monojet #26) as follows:
reana-client restart -w monojet.26 -o FROM=interpretation_stage
This failed without logs (see attached monojet #26.1)
I also tried re-starting from the previous step, make_dict_stage
, and got the same result (see attached monojet #26.2).
I’m not sure if this is a bug or user error. If it looks like a bug, let me know and I can repost it as an issue on the github page repo.